Fig. 2.

Performance of model 1 and model 2. Boxplots displaying the Dice similarity coefficient (DSC) (a), the 95th percentile of the Hausdorff distance (HD95; log applied to improve visualization) (b), and the relative volume difference for whole prostate (RVD (WP)) (c) comparing the performance of model 1, which is trained with set A and model 2, which is trained with set B on the same test set (in-house). The median DSC and HD95 for whole prostate (WP), peripheral zone (PZ), and transition zone (TZ) were 0.933, 0.769, and 0.877 and 5.97, 9.41, and 9.50 mm, respectively, for model 1. The median DSC and HD95 for WP, PZ, and TZ were 0.916, 0.754, and 0.878 and 6.18, 8.88, and 7.44 mm, respectively, for model 2. The median RVD (WP) were 6.66% and − 3.73% for model 1 and model 2, respectively. ns: p ≥ 0.05, **** p < 0.0001